Can We Trust LLM's Logic? Quantifying Uncertainty, Coherence, and Robustness via a Graph-Based Framework

• URL: http://arxiv.org/abs/2607.08017v1
• Date: Thu, 09 Jul 2026 00:45:26 GMT
• Title: Can We Trust LLM's Logic? Quantifying Uncertainty, Coherence, and Robustness via a Graph-Based Framework
• Abstract summary: GRAPHEVAL is a graph-based reasoning framework that re-frames uncertainty quantification (UQ) as a holistic reasoning fidelity problem.<n>We propose a novel UQ metric, Graph Reasoning Coherence Score (GRCS), that quantifies semantic-structural consensus of the reasoning space.<n>We also introduce Graph Self-Consistency (GSC), a medoid-based decoding strategy that trades nominal accuracy for reasoning fidelity.
• Score: 0.6743011393792991
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large-Language Models (LLMs) can be prone to flawed and unfaithful reasoning that decoding strategies like Self-Consistency (SC) fail to detect as they evaluate only final-answer agreement while ignoring the logical validity of intermediate steps. This raises three fundamental questions: How can we reliably quantify uncertainty in LLM reasoning? Can semantic, structural, and causal awareness select more faithful reasoning compared to naïve majority voting? and How robust is reasoning topology under adversarial conditions? To address these questions, we introduce GRAPHEVAL, a graph-based reasoning framework that re-frames uncertainty quantification (UQ) as a holistic reasoning fidelity problem. We propose a novel UQ metric, Graph Reasoning Coherence Score (GRCS), that quantifies semantic-structural consensus of the reasoning space and captures pathological mode collapse and confident hallucinations. We find that GRCS is the only metric that is consistently negatively correlated with reasoning faithfulness across both more capable and smaller models. Additionally, we introduce Graph Self-Consistency (GSC), a medoid-based decoding strategy that trades nominal accuracy for reasoning fidelity, exposing the degree to which SC is inflated by unfaithful lucky guesses in smaller models, while preserving or improving accuracy in more capable ones. Finally, through adversarial medoid ablation, we demonstrate that the GSC-selected path acts as a "load-bearing path" and forcing models away from it degrades reasoning faithfulness and, in targeted cases, causes drops in accuracy.

Related papers

• Beyond Semantic Equivalence: Logical Graphs for LLM Uncertainty Quantification [30.900501335481184]
  We propose Logical Graph Uncertainty (LGU), a framework that explicitly models implication and incompatibility among answers.<n>Across multiple question-answering benchmarks, LGU consistently improves uncertainty estimation over existing methods.
  arXiv  Detail & Related papers  (2026-07-18T16:14:08Z)
• LC-ERD: Mining Latent Logic for Self-Evolving Reasoning via Consistency-Regulated Reward Decomposition [55.572260012037084]
  We introduce LC-ERD (Logic-Consistent Endogenous Reward Decomposition), a framework framing self-alignment as latent structure mining.<n>We derive a Variational Logic Potential by aggregating consensus from the model's Latent Logic Expertise.<n>Experiments show LC-ERD delivers a robust self-evolution path, uncovering trade-offs between logic consistency and accuracy.
  arXiv  Detail & Related papers  (2026-05-19T07:27:50Z)
• Correct Prediction, Wrong Steps? Consensus Reasoning Knowledge Graph for Robust Chain-of-Thought Synthesis [29.865115662967252]
  LLM reasoning traces suffer from complex flaws -- *Step Internal Flaws* (logical errors, hallucinations, etc.) and *Step-wise Flaws* (overthinking, underthinking)
  arXiv  Detail & Related papers  (2026-04-15T17:43:10Z)
• FACT-E: Causality-Inspired Evaluation for Trustworthy Chain-of-Thought Reasoning [49.65751420291115]
  Chain-of-Thought (CoT) prompting has improved LLM reasoning, but models often generate explanations that appear coherent while containing unfaithful intermediate steps.<n>We propose FACT-E, a causality-inspired framework for evaluating CoT quality.<n>We show that FACT-E improves reasoning-trajectory selection and yields stronger in-context learnings.
  arXiv  Detail & Related papers  (2026-04-12T15:35:08Z)
• The Cost of Reasoning: Chain-of-Thought Induces Overconfidence in Vision-Language Models [1.5001933823689926]
  We show that reasoning consistently degrades the quality of most uncertainty estimates.<n>We identify implicit answer conditioning as the primary mechanism.<n>In contrast, agreement-based consistency remains robust and often improves under reasoning.
  arXiv  Detail & Related papers  (2026-03-17T16:12:06Z)
• CausalFlip: A Benchmark for LLM Causal Judgment Beyond Semantic Matching [50.65932158912512]
  We propose a new causal reasoning benchmark, CausalFlip, to encourage the development of new large language models.<n>CaulFlip consists of causal judgment questions built over event triples that could form different confounder, chain, and collider relations.<n>We evaluate LLMs under multiple training paradigms, including answer-only training, explicit Chain-of-Thought supervision, and a proposed internalized causal reasoning approach.
  arXiv  Detail & Related papers  (2026-02-23T18:06:15Z)
• RFEval: Benchmarking Reasoning Faithfulness under Counterfactual Reasoning Intervention in Large Reasoning Models [5.733004743054914]
  Large Reasoning Models (LRMs) exhibit strong performance, yet often produce rationales that sound plausible but fail to reflect their true decision process.<n>We introduce a formal framework for reasoning faithfulness, defined by two testable conditions.<n>We present RFEval, a benchmark of 7,186 instances that probes faithfulness via controlled, output-level counterfactual interventions.
  arXiv  Detail & Related papers  (2026-02-19T03:49:37Z)
• Certainty-Guided Reasoning in Large Language Models: A Dynamic Thinking Budget Approach [0.15749416770494704]
  We show that Certainty-Guided Reasoning (CGR) improves baseline accuracy while reducing token usage.<n>CGR can eliminate millions of tokens in aggregate, with tunable trade-offs between certainty thresholds and efficiency.<n>By integrating confidence into the reasoning process, CGR makes large reasoning language models more adaptive, trustworthy, and resource efficient.
  arXiv  Detail & Related papers  (2025-09-09T14:57:15Z)
• Mitigating Misleading Chain-of-Thought Reasoning with Selective Filtering [59.495717939664246]
  Large language models have manifested remarkable capabilities by leveraging chain-of-thought (CoT) reasoning techniques to solve intricate questions. We propose a novel approach called the selective filtering reasoner (SelF-Reasoner) that assesses the entailment relationship between the question and the candidate reasoning chain. SelF-Reasoner improves the fine-tuned T5 baseline consistently over the ScienceQA, ECQA, and LastLetter tasks.
  arXiv  Detail & Related papers  (2024-03-28T06:28:35Z)
• Language Models with Rationality [57.37201135072838]
  Large language models (LLMs) are proficient at question-answering (QA) It is not always clear how (or even if) an answer follows from their latent "beliefs"
  arXiv  Detail & Related papers  (2023-05-23T17:04:25Z)
• A Weaker Faithfulness Assumption based on Triple Interactions [89.59955143854556]
  We propose a weaker assumption that we call $2$-adjacency faithfulness. We propose a sound orientation rule for causal discovery that applies under weaker assumptions.
  arXiv  Detail & Related papers  (2020-10-27T13:04:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.